Easy Apply
Be an Early Applicant
Generative media platform for developers.
The Role
Design and operate large-scale networks for GPU workloads, automate network infrastructure, ensure network performance, and troubleshoot issues.
Summary Generated by Built In
You are a seasoned networking engineer who has designed, deployed, and operated high-performance networks at scale. At fal, our platform orchestrates AI inference workloads across thousands of GPUs spread over multiple data centers and cloud providers. You will own the network layer that ties it all together—ensuring that model traffic, storage I/O, and control-plane communication are fast, reliable, and secure. You think in terms of packets per second, tail latency, and fabric utilization, and you automate everything you touch.
Key Responsibilities
- Design, build, and operate the network fabric that interconnects our GPU fleet, including spine-leaf architectures, RDMA/RoCEv2 networks for distributed inference, and overlay networks for tenant isolation.
- Own L2/L3 network design across bare-metal and cloud environments, including BGP peering, ECMP, VXLAN/EVPN, and high-bandwidth interconnects between data centers.
- Develop and maintain network automation using Ansible, Terraform, and custom tooling to provision, configure, and validate switches, routers, DPUs, and SmartNICs at scale.
- Instrument deep network observability—build dashboards, alerting, and anomaly detection across our fabric using Prometheus, Grafana, and packet-level telemetry.
- Partner with the Compute and ML Performance teams to tune network paths for AI workloads, minimizing latency for model serving and maximizing throughput for large tensor transfers.
- Drive incident response and root-cause analysis for network-related production issues and build automation to prevent recurrence.
- Evaluate and qualify new networking hardware and software—NICs, switches, DPUs, SONiC, Cumulus, and similar—as we scale to next-generation GPU clusters.
- 8+ years of experience building and operating large-scale networks, ideally in GPU cloud, HPC, or hyperscale environments.
- Deep expertise in Linux networking internals: kernel networking stack, iptables/nftables, tc, eBPF, network namespaces, bonding/teaming, and SR-IOV.
- Strong command of routing and switching protocols: BGP, OSPF, ECMP, VXLAN, EVPN, MPLS, and segment routing.
- Hands-on experience with high-performance networking for AI/ML: RDMA, RoCEv2, InfiniBand, GPUDirect, and NCCL tuning.
- Proficiency automating network infrastructure with Ansible, Python, Go, and Git.
- Experience with network-as-code workflows.
- Familiarity with modern network operating systems such as SONiC, Cumulus Linux, Arista EOS, or Nokia SR Linux.
- Experience with network observability stacks: Prometheus, Grafana, sFlow/NetFlow, and packet capture tools.
- Experience with DPU/SmartNIC programming (NVIDIA BlueField, AMD Pensando) and SDN/NFV architectures.
- Contributions to open-source networking projects (SONiC, FRR, DPDK, eBPF/XDP).
- Experience operating networks that support Kubernetes and container-native workloads (Calico, Cilium, MetalLB).
- Familiarity with data center physical layer design, optics, and cabling at scale
- Interesting and challenging work
- Competitive salary and equity
- A lot of learning and growth opportunities
- We offer visa sponsorship and will help you relocate to San Francisco.
- Health, dental, and vision insurance (US)
- Regular team events and offsite
Remote
Top Skills
Ansible
Arista Eos
Cumulus Linux
Ebpf
Evpn
Git
Go
Gpudirect
Grafana
Infiniband
Linux
Mpls
Network Observability
Prometheus
Python
Rdma
Rocev2
Sonic
Terraform
Vxlan
Am I A Good Fit?
Get Personalized Job Insights.
Our AI-powered fit analysis compares your resume with a job listing so you know if your skills & experience align.
Success! Refresh the page to see how your skills align with this role.
The Company
What We Do
Generative Media Cloud
.png)
